I've checked all the usual suspects for any collation of A, B and E and there are none. There are plenty of versions with extra literary interference but none are really collations.

As you might expect I don't put any store in the Harris fragments at all. They must have been taken from Buchan.

Unless there are other American versions with similar collation of 3 different ballads, 2 of them already concocted, then, as with previous examples we've discussed, the Texas collation must come heavily under suspicion.

Mick, as usual, brilliant! Many thanks.